2003 AP Statistics Form B Scoring Guidelines

AP® Statistics 2003 Scoring Guidelines

Form B

These materials were produced by Educational Testing Service® (ETS®), which develops and administers the examinations of the Advanced Placement Program for the College Board. The College Board and Educational Testing Service (ETS) are dedicated to the principle of equal opportunity, and their

programs, services, and employment policies are guided by that principle.

The College Board is a national nonprofit membership association whose mission is to prepare, inspire, and connect students to college and opportunity. Founded in 1900, the association is composed of more than 4,300 schools, colleges, universities, and other educational organizations. Each year, the College Board serves over three million students and their parents, 22,000 high schools, and 3,500 colleges through major programs and services in college admissions, guidance, assessment, financial aid, enrollment, and teaching and learning. Among its best-known programs are the SAT®, the

PSAT/NMSQT®, and the Advanced Placement Program® (AP®). The College Board is committed to the principles of equity and excellence, and that commitment is embodied in all of its programs, services, activities, and concerns.

For further information, visit www.collegeboard.com

Copyright © 2003 College Entrance Examination Board. All rights reserved. College Board, Advanced Placement Program, AP, AP Vertical Teams, APCD, Pacesetter, Pre-AP, SAT, Student Search Service, and the acorn logo are registered trademarks of the College Entrance Examination Board.

AP Central is a trademark owned by the College Entrance Examination Board. PSAT/NMSQT is a registered trademark jointly owned by the College Entrance Examination Board and the National Merit Scholarship Corporation. Educational Testing Service and ETS are registered trademarks of

Educational Testing Service. Other products and services may be trademarks of their respective owners.

For the College Board’s online home for AP professionals, visit AP Central at apcentral.collegeboard.com.

The materials included in these files are intended for use by AP teachers for course and exam preparation; permission for any other use must be

sought from the Advanced Placement Program®. Teachers may reproduce them, in whole or in part, in limited quantities for noncommercial, face-to-face teaching

purposes. This permission does not apply to any third-party copyrights contained herein. This material may not be mass distributed, electronically or otherwise.

These materials and any copies made of them may not be resold, and the copyright notices must be retained as they appear here.

AP® STATISTICS 2003 SCORING GUIDELINES (Form B)

Copyright © 2003 by College Entrance Examination Board. All rights reserved. Available at apcentral.collegeboard.com.

2

Question 1

Solution Part (a):

The point P does have a large influence on the regression line. When P is removed from the data set, the slope of the line changes from 0.4919 to 0.1500, the intercept changes from 8.107 to 11.123, and the value of R2 drops from 47.6% to 2.5%. Also, the slope is significantly different from 0 when the point P is included in the data set and is not significantly different from 0 when the point P is excluded from the data set.

Part (b):

The regression line for the corrected data will have a negative slope rather than a positive slope, and the intercept would be much larger for the corrected data.

Scoring Part (a) is

Essentially correct (E) if the student

1. identifies the point P as influential AND

2. explains that there have been changes in at least 2 of the following: Slope Statistical significance of the slope Intercept Regression equation Value of R2 (or R2

adj)

OR mentions the change in one of the values above and discusses clearly how point P is extreme in the x direction (if just “extreme,” needs to also explain why that implies the potential for influence).

NOTE: r (0.69 to 0.16) can be mentioned as well, but is not counted separately from R2 unless the student provides clearly distinguishable interpretations of each.



3

Question 1 (cont’d)

Partially correct (P) if the student does one of the following • Identifies Point P as influential but with weak justification (something changes). • Identifies Point P as influential but only one change is noted. • Confuses S as the slope on the computer output and thus states that the point P is not influential since

the slope doesn't change much.

Incorrect (I) if the student only answers “yes” or refers to all of the numbers changing but provides no indication of understanding of what the numbers represent.

Part (b) is

Essentially correct (E) if the student indicates that the sign of the slope would change from positive to negative. The student should explicitly compare the 2 graphs. The student does not need to comment on the change in the intercept.

Partially correct (P) if the student does one of the following • Indicates that the slope will change (including “slope is lower”), but fails to explicitly state that the sign

changes from positive to negative. • Comments only that the value of the correlation changes from positive to negative. • Comments that the line “flattens.”

Incorrect (I) if • Student only comments on the intercept. • Response is very poorly communicated (e.g., “line is negative,” “data are positive,” “data are weak”).

4 Complete Response (EE)

Both parts essentially correct. 3 Substantial Response (EP or PE)

One part essentially correct and the other part partially correct

2 Developing Response (EI or IE or PP) One part essentially correct and the other part incorrect OR Both parts partially correct 1 Minimal Response (PI or IP) One part partially correct



4

Question 2

Solution

Part (a): 89( 31 45) 0.42995207P age − = =

Part (b): 35( 31 45 50,000) 0.3645896P age income over− = =

Part (c):

If annual income and age were independent, the probabilities in (a) and (b) would be equal. Since these probabilities are not equal, annual income and age category are not independent for adults in this sample.

Scoring Part (a) is scored as either essentially correct (E) (may be minor arithmetic errors) or incorrect (I). Part (b) is

Essentially correct (E) if the conditional probability is correctly calculated.

Partially correct (P) if the student reverses the conditioning, calculating

35( 50,000 31 45) 0.393389

P income over age − = =

OR

calculates the correct probability for the wrong column, e.g., 3264

Incorrect (I) if the student calculates the joint probability: 35 0.169207

=

Part (c) is

Essentially correct (E) if the student 1. indicates that the two variables are not independent

AND 2. the explanation is tied to the fact that the probabilities in parts (a) and (b) are not equal (the answer

must be based on parts (a) and (b))



5


Partially correct (P) if the student indicates that the two variables are not independent, but the explanation is incorrect, or is not based on the answers to parts (a) and (b); i.e., performing new correct calculations instead of referring to those in parts (a) and (b). For example: determining the probability of the intersection and comparing to the two individual probabilities

35 96 890.169, which does not equal (0.43)(0.46) ,207 207 207

� �= ⋅ =� �

reversing conditions

96 35e.g., 0.464, which does not equal 0.393 ,207 89

� �= =� �

or other conditional probability comparisons.

Incorrect (I) if the student fails to give a numerical justification to support the argument.

OR Incorrect if the student does one of the following • performs an incorrect additional calculation • says the variables are independent based entirely on the context. • performs a chi-square test ( 2 5.38,� = p-value = 0.496) since this addresses independence in the

population instead of the sample • only states “yes, independent” with no justification

NOTE: If either of the probabilities calculated in (a) or (b) are incorrect, part (c) should be scored as if those probabilities were correct. For example, if the student incorrectly calculated the same answer for parts (a) and (b), part (c) would be scored as correct if the student states that you can't tell if the two variables are independent because you would need to check all age-gender combinations. 4 Complete Response (EEE)

All three parts essentially correct

3 Substantial Response (EEP, EPE, EPP, IEE)

Part (a) essentially correct and parts (b) and (c) at least partially correct OR

Part (a) incorrect and parts (b) and (c) essentially correct

2 Developing Response (EEI, EIE, EPI, EIP, IEP, IPE, IPP)

Part (a) essentially correct and one (but not both) of parts (b) and (c) correct OR

Part (a) incorrect and both parts (b) and (c) at least partially correct

1 Minimal Response (EII, IPI, IIP, IEI, IIE)

Part (a) essentially correct and parts (b) and (c) incorrect OR

Part (a) incorrect and one of parts (b) and (c) partially correct



6

Question 3

Solution Part (a):

This study is an experiment. The researchers imposed treatments and subjects were randomly assigned to the two treatment groups.

Part (b):

The two-proportion z test could be used to compare the proportion of volunteers who get the flu for the two conditions. The hypotheses would be 0 aH : 0 H : 0T C T Cp p versus p p− = − <

OR 0 aH : 0 H : 0C T C Tp p versus p p− = − >

OR 0 aH : H :T C T Cp p versus p p= <

OR 0 aH : H :T C C Tp p versus p p= >

where Tp is the proportion of those receiving vitamin C (from the population of students who would

volunteer for such a study) who contract the flu (or the probability that such a student receiving vitamin C contracts the flu).

and Cp is the proportion of those receiving placebo (from the population of all students who would

volunteer for such a study) who contract the flu.

Note: Could also define p as the proportion who do not contract the flu and then reverse the direction of the alternative hypothesis statement.



7


Scoring This problem has 3 elements; part (a) is one element and part (b) is divided into two elements (naming the test and stating the hypotheses). Each element is scored as either essentially correct (E), partially correct (P), or incorrect (I).

Element 1 (part (a)) is

Essentially correct (E) if the student

1. concludes that the study is an experiment AND

2. the explanation is tied to the fact that the researchers imposed a treatment (controlled which medicine) OR

states that there was random assignment of subjects to treatments (random division into two groups)

Partially correct (P) if the student indicates that the study is an experiment, but the explanation is missing or does not include either that the researcher imposes treatments or that there is random assignment of subjects to treatments (e.g., “experimenters controlled factors”). Incorrect (I) if the student says one of the following: • this is not an experiment because the study used volunteers or because the subjects were not randomly

selected. • this is not an experiment because the experimenters did not control everything

Element 2 (naming the test) is Essentially correct (E) if the student identifies the two-sample z test for proportions (all 3 parts of the name are needed)

Partially correct (P) if the student does either of the following • identifies a chi-square test (homogeneity) • only gives two components of the name (e.g., “2 sample z test”)

NOTE: If student says “two sample p test,” grade holistically. Incorrect (I) if the student identifies any test involving means



8


Element 3 (stating the hypotheses) is

Essentially correct (E) if the student does any of the following • gives a correct pair of one-sided hypotheses. If the student uses Tp and Cp (for treatment and

control) or Cp and pP (for vitamin C and placebo), they need not define the parameters to get credit for this element. If they use p1 and p2, they must identify which is treatment and which is control.

• states correct hypotheses for the test described in element 2 (e.g., a two-sided alternative hypothesis or verbal description for the chi-square test, or involving two means for a two-sample t test).

• provides correct statement of null and alternative hypotheses in words. Partially correct (P) if either

• the direction of the inequality in the alternative hypothesis is incorrect, or if a two sided alternative is specified, or

• p1 and p2 are used in the hypotheses without specifying which refers to the treatment group and which refers to the control group

Incorrect (I) if the student does any of the following

• specifies sample proportions in the hypotheses (unless defined as parameters). • uses proportions in the hypotheses and means in the test procedure or vice versa. • reverses the hypotheses.

NOTE: Elements 2 and 3 could be correct if a one-sample test for a proportion is specified and the hypotheses are given as

0 0 a 0H : H :T Tp p versus p p= < and 0p is defined to be the true proportion in the population of untreated volunteers

who get the flu. But to receive credit for this solution 0p must be correctly defined.



9


4 Complete Response (3E)


3 Substantial Response (2E 1P)

Two parts essentially correct and 1 part partially correct

2 Developing Response (2E 0P or 1E 2P or 3P)

2 parts essentially correct and no parts partially correct OR

One part essentially correct and 2 parts partially correct OR

3 parts partially correct 1 Minimal Response (1E 1P or 1E 0P or 0E 2P)

One part essentially correct and either 0 or 1 parts partially correct OR

No parts essentially correct and 2 parts partially correct



10

Question 4

Solution Part (a):

Assign each subject a number from 001 to 300 and then use a random number table or a random number generator to select 150 of the 300 for the new filter group. The other 150 would be assigned to the standard filter group.

OR For each subject, flip a coin. If the coin lands H, assign the subject to the new filter group; otherwise assign the subject to the standard filter group. Continue in this way until one of the groups has 150 subjects. Assign all remaining subjects to the other group.

Part (b):

Without a comparison group, the cholesterol level could change overall, but we would not be able to determine whether the observed change was due to some other extraneous variable that changed during the 10-week period. For example, diet might change with time of the year, and the diet might result in changes in cholesterol changes. So a change in cholesterol would not be attributable to the new coffee filter. The addition of a control group enables the researchers to assess the mean change in cholesterol level due to the coffee filter, as opposed to just determining if the cholesterol level changed. The control group eliminates the confounding variable of another change that might have occurred over the 10-week period.

Part (c):

The two-sample t test for means or mean differences would be used (or the two-sample z test for means).

Part (d):

If it is known that smoking is related to changes in cholesterol level, it would be best to control for smoking by using only nonsmokers. This eliminates smoking as a source of variability, creating more homogenous groups, enabling more direct comparisons between the treatment and control groups and more precise estimates of the treatment effects (though we will only be able to generalize the results to nonsmokers).



11


Scoring Part (a) is

Essentially correct (E) if the student describes a method that is 1. based on random assignment and

2. will result in equal numbers of subjects in each group

Partially correct (P) if the student does any of the following �� describes a method based on random assignment, but that does not ensure an equal number of

subjects in each group. �� attempts to describe a method of random assignment that ensures 150 subjects in each group, but

the explanation of random assignment is not clear or incomplete. This may include assigning numbers to the subjects and selecting the even numbered subjects for the treatment group, if it is clear the student believes this will randomize the groups.

Incorrect (I) if the student describes any method not based on random assignment. For example, allows the subjects to self-select themselves into the 2 groups, or just referring to “random assignment” but not describing the method of assignment.

Part (b) is

Essentially correct (E) if the student 1. describes the need for a comparison group and

2. explains how the control group allows the researchers to attribute the change in cholesterol to the new filter as opposed to natural variability or another confounding variable (that a simple before-after measurement would not be sufficient to conclude that the change is due to the new filter). OR discusses “confounding” or “lurking” variables and describes the variable in such a way that it could lead

to the change in the cholesterol levels during the 10 week time period for these subjects.

Partially correct (P) if the student does one of the following

�� indicates that the inclusion of a control group allows the new and standard filters to be compared, but the explanation does not adequately explain the need for a comparison group to control for other changes during the 10 week period.

�� refers to controlling for “confounding or lurking variables” or “factors,” and describes a reasonable

variable that could change cholesterol but does not discuss the need for comparison with the control group to eliminate these factors as potential explanations for the change in cholesterol with the new filter.

Incorrect (I) if the student fails to indicate the need for a comparison group and does not describe a good

confounding variable in context.



12


Parts (c) and (d) are scored together. These are essentially correct (E) if both parts (c) and (d) are answered correctly. Part (cd) is partially correct (P) if only one of parts (c) or (d) is answered correctly. Part (c) is

Correct if the student indicates a procedure to compare means for two independent samples. The student can refer to a two sample t test.

Incorrect if the student states only “two sample z test” or any other test procedure.

Part (d) is

Correct if the student explains how smoking could be related to initial cholesterol levels at the start of the study (there is something different about smokers) and indicates a desire for more homogenous groups or recognizes that focusing only on nonsmokers will reduce variability (akin to blocking). Student may use the word “confounding” if (through explanation) they indicate they intended an extraneous variable. Incorrect if the student does either of the following

��Describes smoking as a confounding variable. ��Only states that smoking is related to cholesterol.

4 Complete Response (3E)


3 Substantial Response (2E 1P)

2 Developing Response (2E 0P or 1E 2P or 3P)

1 Minimal Response (1E 1P or 1E 0P or 0E 2P) NOTE: A response with 1E and 1P can be graded holistically based on the strength of the responses in parts (b) and (d).



13

Question 5

Solution Part (a):

P(a number on all 3 spins) = [P(number)]3 since the outcomes are independent from spin to spin

= 33 0.4219

4� � =� �

Part (b):

Winnings 0 900 1000 1300 Probability 0.25 0.25 0.25 0.25 E(winnings) = = 0(0.25)+900(0.25)+1000(0.25)+1300(0.25) = 800i ix p�

Or E(winnings on 4th spin) = -800(0.25) + 100(0.25) + 200(0.25) + 500(0.25) = 0 So E(winnings) = initial amount + E(winnings on 4th spin) = 800 + 0 = 800

Part (c): Element 1: States a correct pair of hypotheses

H0: The four outcomes are equally likely 1 2 3 41or 4

p p p p� �= = = =� �

Ha: The four outcomes are not equally likely (or at least one pi differs from 14

)

Element 2: Identifies a correct test (by name or by formula) and checks appropriate conditions.

Chi-square test (for goodness of fit) 2

2 ( )Obs ExpExp

�−

=�

Conditions: Outcomes of spins of the wheel are independent and large sample size.

The problem states that successive spins of the wheel are independent.

The expected counts are all equal to 25, which is greater than 5 (or 10), so the sample size is large enough to proceed.



14


Element 3: Correct mechanics, including the value of the test statistic, df, and p-value (or rejection region)

Expected counts: 25 for each of the 4 cells

2 2 2

2 ( ) (33 25) (26 25) 4.2425 25

Obs ExpExp

�− − −= = + + =� �

df = 4-1 = 3 p-value = .2367 (from tables p-value > 0.10, from Graphing Calculator: p-value = 0.23669, from table rejection region for 0.05� = is 7.81, 0.01� = is 11.34)

Element 4: Using the results of the statistical test, states a correct conclusion in the context of the problem.

Because the p-value is greater than the stated � (or because the p-value is large, or because the test statistic does not fall in the rejection region), fail to reject H0. There is not convincing evidence that the four outcomes on the wheel are not equally likely. That is, we don’t have convincing evidence against the conjecture that the four outcomes on the wheel are equally likely.

Scoring

Part (a): 1 for correct answer (including Binomial calculation) = 0.4219

12

if answer is 34

or 31 0.0156

4� � =� �

or 31 (3) 0.047

4� � =� �

or 0.4219 with no work

Part (b): 1 if the expected value, 800, is correct (except for minor computational errors)

12

if expected value is computed as

�� 800 + expected winnings on one spin, or 800 + 200 = 1000 ��E(outcome on one spin)=200 but then solution breaks down ��E(winnings on 4th spin)=0 but then solution breaks down

��With fairly major computational errors 3200e.g., 3

� ��



15


0 if

�� answer of 800 is given but no work is shown or bad logic, e.g., 4(200) �� expected value formula is given but no calculations are done �� outcomes are set up correctly but no expected value is calculated

Part (c): 12

for each element of the test that is correct

1. statement of hypotheses 2. identification of test and check of sample size condition 3. correct mechanics 2 4.24� = , df=3, p-value=0.2367, and/or 2

3,0.05 7.81� = 4. statement of conclusion (fail to reject)

If both an α and a p-value are given, the linkage is implied. If no α is given, the solution must be explicit about the linkage by giving a correct interpretation of the p-value or explaining how the conclusion follows from the p-value.

NOTE: If the p-value in element 3 is incorrect but the conclusion is consistent with the computed p-value, element 4 can be considered as correct.

4 Complete Response

Score of 4 from parts (a) through (c)

3 Substantial Response Score of 3 from parts (a) through (c)

2 Developing Response

Score of 2 from parts (a) through (c)

1 Minimal Response Score of 1 from parts (a) through (c)

IF A PAPER IS BETWEEN TWO SCORES (FOR EXAMPLE, 2 ½ PARTS) USE A HOLISTIC APPROACH TO DETERMINE WHETHER TO SCORE UP OR DOWN DEPENDING ON THE STRENGTH OF THE RESPONSE AND COMMUNICATION.



16

Question 6

Solution Part (a): Part (a) is scored based on four elements Element 1: Identifies appropriate confidence interval by name or by formula and checks the appropriate conditions.

One sample confidence interval for a proportion ˆ ˆ(1 )ˆ * p pp z

n−

±

Conditions: random sample from a large population and large sample size. Since the sample of size 2000 was a random sample from the population of all patients at the HMO, it is reasonable to consider the sample of 40 to 44 year old females (n=370) as a random sample of the 40 to 44 year old female patients of the HMO.

ˆ ˆ ˆ0.10 370(0.10) 37 (1 ) 370(0.90) 333p np n p= = = − = = Since both ˆnp and ˆ(1 )n p− are both greater than 5 (or 10), the sample size is large enough to proceed.

Since these conditions are met, if is reasonable to proceed with the following calculations and interpretations.

Element 2: Correct mechanics

ˆ ˆ(1 ) (0.10)(0.90)ˆ * 0.10 1.96 0.10 0.03 (0.07,0.13)

370p pp z

n−

± = ± = ± =

Graphing calculator: (0.06943, 0.13057)

Element 3: Interpretation of the interval

We can be 95% confident that the true proportion of this HMO’s 40 to 44 year old female patients who contracted the disease is between 0.07 and 0.13.



17


Element 4: Interpretation of the confidence level

Ninety five percent of all possible random samples of size 370 from this population will result in a confidence interval that includes the true population proportion of this HMO’s 40 to 44 year old female patients who contracted the disease. Can also say, approximately 95% of a large number of intervals will contain the population proportion.

OR The method used to produce this interval will fail to capture the population proportion about 5% of the time.

Part (b):

The width of the interval is determined by the magnitude of ˆ ˆ(1 ) ,p p

n− which depends on p̂ and n. If the

sample proportions are equal, then the confidence interval widths will be the same if the sample sizes are the same for all 8 age-gender groups. Thus, we need to take random samples of size 250 from each of the 8 groups.

Part (c):

The width of the interval is proportional to ˆ ˆ(1 ) .p p

n− The interval widths for all of the groups will be the

same if ˆ ˆ(1 )p p

n− is the same for each group. This will happen when the sample size is proportional

to ˆ ˆ(1 ).p p− For this to happen, we need (approximately)

1 2 3 4

(0.05)(0.95) (0.08)(0.92) (0.20)(0.80) (0.35)(0.65)n n n n

= = =

where 1 2 3 4 2000n n n n+ + + =

Solving these equations we get Age Group Sample Size (1 )p p

n−

35 to 39 187 0.0159 40 to 44 289 0.0159 45 to 49 629 0.0159 50 to 54 895 0.0159



18


OR

it is sufficient to say that since ˆ ˆ(1 )p p−� is anticipated to be

ˆ ˆ(1 ) (0.05)(0.95) (0.08)(0.92) (0.20)(0.80) (0.35)(0.65) 0.0475 0.0736 0.1600 0.2275p p− = + + + = + + +�

0.5086= we want

10.0475 2000 186.790.5086

n � �= =� �

20.0736 2000 289.420.5086

n � �= =� �

30.1600 2000 629.180.5086

n � �= =� �

40.2275 2000 894.610.5086

n � �= =� �

Scoring Part (a) is scored based on the number of the four elements that are correct, ½ point for each element.

1. Naming procedure and checking conditions (It is OK not to repeat SRS since stated given in the problem statement.)

2. Carrying out the mechanics for a 95% confidence interval (0.0694, 0.1306) 3. Interpreting the confidence interval in context 4. Interpreting the confidence level

Note: Parts 3 and 4 need to be read together (correct interpretations in reverse order should be graded as essentially correct). Common errors in parts 3 and 4 include: No context in part 3, claiming 95 out of 100 intervals exactly, or describing the interval for a population mean (this last error will only count once against element 3 and element 4).

Parts (b) and (c) are scored as either essentially correct (E), partially correct (P), or incorrect (I). Score essentially

correct as 1 point and partially correct as 12

point.

Part (b) is

Essentially correct (E) if the student 1. specifies equal sample sizes of 250. (If an explanation clearly indicates that all samples will be 250, without specifying the actual number, credit will be awarded.)

AND

2. provides justification that appeals to the width of the interval being dependent on ˆ ˆ(1 )p p

n−



19


Partially correct (P) if the student does one of the following

• specifies that equal samples sizes of 250 are needed, but does not justify the answer based

onˆ ˆ(1 )p p

n− .

• justifies that the width of the interval depends on n (through standard error), but fails to state that the sample size is 250 for all 8 groups.

• justifies that the sample sizes need to be equal through the standard error but does not focus on all eight groups. For example: considers either just males and females (1000/1000) or equal gender in given age groups (398/300/177/105), or equal samples sizes in the four age groups (500/500/500/500).

Incorrect (I) if the answer

• Specifies that equal sample sizes are needed but gives with no explanation. Part (c) is

Essentially correct (E) if 1. the explanation recognizes that the sample sizes should be proportional to p(1-p) AND 2. the resulting sample sizes are computed OR

conditions that would allow for the computation of the sample sizes are given (setting up the equations and noting that the sample sizes must sum to 2000). It is not necessary to actually compute the sample sizes.

Partially correct (P) if the student • correctly computes the sample sizes but the justification is missing or incorrect OR • chooses the samples sizes to be proportional to p (rather than p(1-p)), resulting in sample sizes of 147,

235, 588, 1029. Note, 0.05 (2000) 1470.68

= etc.

OR • states that the sample sizes depend on n and on p̂ and ˆ ˆ(1 )p p− , but fails to indicate how to carry out

the calculation. For example, trying to set the confidence intervals equal to each other and noting that the sum of the n’s is 2000.

NOTE: If a student incorrectly solves their equations but obtains plausible answers (between 0 and 2000) then score as essentially correct.

Incorrect (I) if equal sample sizes are recommended.



20


4 Complete Response

Score of 4 on parts (a) through (c)

3 Substantial Response Score of 3 on parts (a) through (c)

2 Developing Response

Score of 2 on parts (a) through (c)

1 Minimal Response Score of 1 on parts (a) through (c)

IF A PAPER IS BETWEEN TWO SCORES (FOR EXAMPLE, 2 ½) USE A HOLISTIC APPROACH TO DETERMINE WHETHER TO SCORE UP OR DOWN DEPENDING ON THE STRENGTH OF THE RESPONSE AND COMMUNICATION.

2003 AP Statistics Form B Scoring Guidelines

Documents

Transcript of 2003 AP Statistics Form B Scoring Guidelines